SplatChat Android App

SplatChat (formerly SnapConnect) is a mobile social media application designed for users to share moments with friends through images and videos in real-time. The core concept revolves around ephemeral content, where shared media is automatically deleted after 24 hours.

Features

Core Features

• User Authentication: Sign up and login with email/password
• Friends Management: Add friends, manage friend requests, view friends list
• Real-time Image and Video Sharing: Share photos and videos with friends
• Ephemeral Content (Stories): Posts automatically delete after 24 hours
• Comments: Comment on friends' stories with real-time updates
• Group Messaging: Create group chats and share content
• 📸 Camera Integration - Capture photos and videos with custom filters
• 📱 Stories - Share ephemeral content that disappears after 24 hours
• 💬 Real-time Messaging - Chat with friends using Supabase real-time subscriptions
• 🎭 AR Filters - Apply face filters using ML Kit face detection
• 🎨 Material Design 3 - Modern UI with dynamic theming

Advanced Features

• Story Reactions: Like and dislike stories with real-time counts
• User Score System: Profile displays total score (likes minus dislikes across all stories)
• Story Auto-Advance: Stories auto-advance after 8 seconds with toggle control
• Interactive Story Viewer: Progress bars, manual navigation, pause functionality
• Tutorial System: Guided onboarding for new users
• Privacy Controls: Granular permission management for camera, notifications, storage
• AI-Powered Inspiration:
  • AI-generated captions and mood boards
  • Style analysis and tagging
  • Style-based story discovery
• Media Preview: Preview and edit content before posting
• Profile Management:
  • Avatar upload and editing
  • Profile stats (stories, friends, score)
  • Display name and username customization
• Style Gallery: Browse and discover stories by artistic style
• Notification Settings: Customizable notification preferences
• Story Privacy: Set stories as public or friends-only
• Enhanced Video Support: Full video playback with auto-advance control

Tech Stack

• Frontend: Android (Kotlin) with Jetpack Compose
• Architecture: MVVM (Model-View-ViewModel)
• Backend: Supabase (Auth, Database, Storage, Realtime)
• Dependency Injection: Hilt
• Navigation: Navigation Compose
• Image Loading: Coil
• Camera: CameraX

Setup Instructions

Prerequisites

• Android Studio Arctic Fox or later
• JDK 17 or later
• Android SDK 34
• Minimum Android device/emulator with API 24 (Android 7.0)

Supabase Setup

1. Create a Supabase project at supabase.com

2. Get your Supabase URL and Anon Key from the project settings

3. Run the database setup script in your Supabase SQL editor:

   • Execute the SQL in supabase_setup.sql to create tables and policies

4. Update the Supabase configuration in app/src/main/java/com/example/snapconnect/di/AppModule.kt:

// Replace with your Supabase project details
private const val SUPABASE_URL = "YOUR_SUPABASE_URL"
private const val SUPABASE_ANON_KEY = "YOUR_SUPABASE_ANON_KEY"

Android App Setup

1. Clone the repository
2. Open the project in Android Studio
3. Sync the project and build
4. Run on device or emulator

Project Structure

app/
├── src/main/java/com/example/snapconnect/
│   ├── data/
│   │   ├── model/          # Data models
│   │   └── repository/     # Repository classes
│   ├── di/                 # Dependency injection modules
│   ├── navigation/         # Navigation setup
│   ├── ui/
│   │   ├── screens/        # Composable screens
│   │   └── theme/          # App theming
│   ├── MainActivity.kt
│   └── SnapConnectApplication.kt
└── src/main/res/          # Resources

Current Implementation Status

✅ Completed:

• Project setup with Jetpack Compose
• Dependency injection with Hilt
• Complete navigation structure with all screens
• Authentication system (Login/SignUp) with Supabase
• Main app structure with bottom navigation
• Data models for all features
• Theme and styling with SplatChat branding
• Camera functionality with filters and AR effects
• Story system with posting, viewing, and auto-advance
• Real-time messaging and group chats
• Friends management with requests and social features
• Story reactions (likes/dislikes) with real-time updates
• User profiles with stats, avatars, and editing
• Privacy controls and permission management
• AI integration for captions, style analysis, and inspiration
• Tutorial system for onboarding
• Media storage and retrieval via Supabase
• Story privacy controls (public/friends-only)
• Notification system and settings

Development Notes

• The app uses Material 3 design system with SplatChat branding
• All screens follow MVVM architecture pattern with Hilt dependency injection
• Supabase handles authentication, database, storage, and real-time features
• Camera and microphone permissions required for full functionality
• Story auto-advance timing: 8 seconds for images, video length for videos
• AI features require OpenAI API key and Pinecone vector database
• Row Level Security (RLS) policies enforce privacy controls

License

This project is for educational purposes.

Backend Micro-Service (RAG)

A lightweight FastAPI service (snapconnect-backend/) now powers the Inspiration & Reference Engine.

Key endpoints (protected by X-API-Key)

Method	Path	Description
POST	/api/v1/inspiration/moodboard	Generate mood board from a text prompt
POST	/api/v1/inspiration/style-analysis	Analyse artistic style from caption text

See the backend README for setup, seed scripts, and deployment notes.

Recent Updates

Story Features

• Stories can be set as Public (viewable by anyone) or Private (friends only)
• Story Reactions: Users can like or dislike stories with real-time count updates
• Auto-advance: Stories advance automatically after 8 seconds (toggleable)
• Interactive Viewer: Progress bars, manual navigation (tap left/right), pause on hold
• Enhanced Media Support: Full image and video support with optimized playback
• Privacy controls enforced via Row Level Security (RLS) policies in Supabase

Group Chat Support

• Create group chats with multiple friends
• New "Create Group" screen accessible from Messages tab
• Groups stored in groups table with member_ids array
• Direct messages and group chats share the same messaging infrastructure

AI-Powered Style Detection

• Automatic style tagging for stories using OpenAI embeddings + Pinecone vector search
• Supported styles include: pop-art, watercolor, cyberpunk, minimalist, and more
• Pipeline: Story creation → AI caption generation → Style detection → Vector storage
• Style tags stored in style_tags array column
• AI-generated captions stored in ai_caption column

Backend Infrastructure

• FastAPI backend deployed on Render
• Pinecone vector database (1024-dimensional, serverless)
• LangChain integration for RAG (Retrieval Augmented Generation)
• Style taxonomy namespace for style similarity search

Backend Configuration

Required Environment Variables

# OpenAI
OPENAI_API_KEY=your-openai-api-key

# Pinecone Vector Database
PINECONE_API_KEY=your-pinecone-api-key
PINECONE_ENVIRONMENT=us-east-1-aws  # Your index region
PINECONE_INDEX_NAME=snapconnect     # 1024-dim index

# Supabase
SUPABASE_URL=your-supabase-url
SUPABASE_SERVICE_ROLE_KEY=your-service-role-key

User Experience Features

• Tutorial System: Interactive onboarding for new users
• Privacy Screen: Comprehensive permission management for camera, notifications, storage
• Profile Management: Avatar upload, profile editing, user statistics
• User Score: Dynamic score calculation (total likes - total dislikes)
• Notification Settings: Granular control over app notifications
• Style Discovery: Browse stories by AI-detected artistic styles

Database Schema Updates

-- Stories table with full feature set
ALTER TABLE stories 
  ADD COLUMN is_public BOOLEAN DEFAULT true,
  ADD COLUMN style_tags TEXT[] DEFAULT '{}',
  ADD COLUMN ai_caption TEXT,
  ADD COLUMN likes_count INTEGER DEFAULT 0,
  ADD COLUMN dislikes_count INTEGER DEFAULT 0;

-- Story reactions table
CREATE TABLE story_reactions (
    id UUID DEFAULT gen_random_uuid() PRIMARY KEY,
    story_id UUID REFERENCES stories(id) ON DELETE CASCADE,
    user_id UUID REFERENCES auth.users(id) ON DELETE CASCADE,
    reaction_type TEXT CHECK (reaction_type IN ('LIKE', 'DISLIKE')),
    created_at TIMESTAMPTZ DEFAULT NOW(),
    UNIQUE(story_id, user_id)
);

-- Row Level Security for private stories
CREATE POLICY select_stories_with_privacy
ON stories
FOR SELECT
USING (
    is_public
    OR user_id = auth.uid()
    OR EXISTS (
        SELECT 1
        FROM friends f
        WHERE f.status = 'ACCEPTED'
        AND (
            (f.user_id = auth.uid() AND f.friend_id = stories.user_id)
            OR (f.friend_id = auth.uid() AND f.user_id = stories.user_id)
        )
    )
);